# Can I use IRLS (e.g., Lp-norm) regularization on the gravity and magnetic models while still using cross-gradient as a coupling term?

**URL:** <https://simpeg.discourse.group/t/can-i-use-irls-e-g-lp-norm-regularization-on-the-gravity-and-magnetic-models-while-still-using-cross-gradient-as-a-coupling-term/694>\
**Category:** Potential Fields\
**Created:** [April 3, 2025, 1:31pm UTC](https://simpeg.discourse.group/t/can-i-use-irls-e-g-lp-norm-regularization-on-the-gravity-and-magnetic-models-while-still-using-cross-gradient-as-a-coupling-term/694 "2025-04-03T13:31:59Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![Mr.Xingfeng](https://avatars.discourse-cdn.com/v4/letter/m/cab0a1/32.png) [@Mr.Xingfeng](https://simpeg.discourse.group/u/Mr.Xingfeng)\
**Post date:** [April 3, 2025, 1:31pm UTC](https://simpeg.discourse.group/t/can-i-use-irls-e-g-lp-norm-regularization-on-the-gravity-and-magnetic-models-while-still-using-cross-gradient-as-a-coupling-term/694/1 "2025-04-03T13:31:59Z")

</div>

I’m working on a joint inversion of gravity and magnetic data in SimPEG, and I want to incorporate cross-gradient coupling between the two models.

I know that SimPEG supports IRLS for sparsity-promoting regularization via Update\_IRLS,Can I use IRLS (e.g., Lp-norm) regularization on the gravity and magnetic models while still using cross-gradient as a coupling term? Additionally, both ‘Update\_IRLS’ and ‘PairedBetaSchedule’ contain parameters named ‘cooling\_factor’ and ‘cooling\_rate’.  
Do these overlap or conflict in any way when used together in a joint inversion?